New Understanding Of Neural Circuits May Help Speed Development Of
Thought-controlled Prosthetic Devices For Paralyzed

Source:  Society For Neuroscience

Date:  2004-11-03

You can read this article, and more generally, function in the world,
thanks to precise communications between circuits of millions of
nerve cells in your brain and spinal cord. Once a mystery, new
studies now decipher some of this complex chatter.

“The clinical implications of this type of basic science research are
enormous,” says György Buzsáki, MD, PhD, of Rutgers University , an
expert on neural circuits. “By understanding the communication
process of normal neural circuits, we can determine ways to help
those who experience an injury or have a disorder that has damaged
their circuits and impaired their function.”

For example, the research will likely aid the development of “thought-
controlled prosthetic devices,” which could help amputees or those
with paralysis who can't use their limbs. Normally circuits of nerve
cells in the brain constantly compute and transmit complex signals
through the spinal cord and out peripheral nerves in order to move an
arm, hand, leg, or foot. An injury that causes the loss of a limb or
damage to the spinal cord can sever this communication line and
prevent desired movement. In theory, a thought-controlled prosthetic
device, also known as a brain-machine interface, could bypass the
loss or damage by directly interpreting nerve cell signals and then
launching movement in a prosthetic limb.

The development of these devices requires a two-pronged, parallel
approach. “The first is the understanding of the self-organized
nature of healthy neural circuits,” says Buzsáki. “Second is the use
of this information to control prosthetic devices and feed back
information from them to the brain.” Researchers will report the
latest advances in device development in the press conference on
prosthetic devices, at the Society For Neuroscience annual meeting.
This session will focus on the first issue and detail the progress in
the basic science fields that help decipher neural circuit
communication.

One new analysis that decoded some circuit activity may help simplify
the development process of a thought-controlled prosthetic device.
“Our findings represent a significant step toward the development of
thought-controlled prosthetic devices, such as robotic arms for
patients with severe neurological injuries, including those who
suffer from complete paralysis,” says study author Miguel Nicolelis,
MD, PhD, of Duke University.

In the study, the researchers examined 11 awake patients who were
undergoing surgery to treat the brain disorder Parkinson's disease.
Before their surgical treatment was performed the researchers used
electrodes to simultaneously record signals from groups of nerve
cells in two regions of the brain that aid movement control. Nerve
cells talk to each other by sending electrical signals. One common
way to study this activity is to place tiny wires, or electrodes, in
the brain. This allows researchers to “see” the electrical signals
from individual neurons.

During the recordings, the patients performed a visual-feedback hand-
gripping task. “Remarkably, small groups of 3 to 35 simultaneously
recorded nerve cells were sufficiently information-rich to predict
gripping force during a 2.5-minute test period with considerable
accuracy,” says Nicolelis. “This finding could simplify the
development process of thought-controlled prosthetic devices.”

A thought-controlled prosthetic device requires a computer program
that can read and decode the complex language of nerve cells in order
to set off an intended movement in a robotic limb. “Our study
suggests that the computer program may only need to understand the
language of just a few hundred randomly selected small groups of
cells to drive the robotic limbs,” says Nicolelis.

Another new study from the Nicolelis laboratory at Duke also may help
simplify the development of computer programs for thought-controlled
prosthetic devices. “We found that certain common components seen in
the signals from groups of nerve cells better predict a variety of
movement directions than the information from an individual nerve
cell,” says Nicolelis.

In the study, researchers implanted signal-recording electrodes into
brain areas known to aid reaching movements in two monkeys. Then Duke
researchers trained the implanted monkeys to control a cursor on a
computer screen using a joystick. Targets appeared on the screen, and
the monkeys reached for them with the cursor. In the next series of
experiments the monkeys controlled movements of a robot arm using the
same joystick. The robot was invisible to the monkeys, but the cursor
provided visual feedback of the robot's position.

An analysis determined that common components from the signaling
activity of a large population of nerve cells better predicted
movement directions than activity from individual nerve cells. “We
believe that we can use the information from the large cell
populations to make the robot arm perform a variety of movement
tasks,” says Nicolelis. “Together the research may help simplify the
development of the computer program portion of thought-controlled
prosthetic devices.”

Other researchers at the Massachusetts Institute of Technology
developed two new techniques that will help scientists better analyze
the activity of neural circuits.

“The combination of new electrical and optical techniques will give
us unprecedented opportunities to study how circuits in the brain
work,” says study author Michale Fee, PhD.

On the electrical side, Fee and his colleagues developed a method
that improves the standard use of electrodes to measure the
electrical signals between nerve cells. “We have made a tiny device,
called a microdrive, which holds up to three electrodes at a precise
position in select brain areas,” says Fee. “In addition, tiny motors
in the device allow us to move the electrodes around by remote
control, find electrical signals from individual nerve cells, and
then record their patterns of activity.”

This setup is ideal for examining animals, like songbirds, during
their normal behaviors. “We were able to position the electrodes and
find nerve cells in the songbird's brain without bothering the bird
at all,” says Fee. “Because of this ability, we discovered that a
particular group of nerve cells control the precise timing of the
bird's songs.”

Another way to study nerve cell activity in the brain is to look at
cells under a microscope. When nerve cells signal each other
electrically, chemical changes happen inside the cell. For example,
electrical activity in a nerve cell causes tiny amounts of calcium
ions to flow into the cell, and optical techniques can highlight this
change. “There are particular fluorescent dyes that change the amount
of light they give off depending on how much calcium is around,” says
Fee. “So if we put some of this dye inside a nerve cell, we can see
the cell flash as it does its work and signals other nerve cells.”
But in the past, this technique couldn't be used for moving animals.

Fee and his colleagues have changed this by developing a special
miniature microscope that can be placed on the head of a small animal
like a rat. “This allows us to watch changes in fluorescence deep in
the brain in an animal that is free to move around,” says Fee. “In
the future, we should be able to watch whole groups of nerve cells in
the brain flashing their signals to each other as an animal goes
about its business.”

Editor's Note: The original news release can be found here.
http://web.sfn.org/content/AboutSFN1/NewsReleases/am2004_neural.html

This story has been adapted from a news release issued by Society For
Neuroscience.

SOURCE: Science Daily (press release)
http://tinyurl.com/5u7p7

* * *
Murray Charters <[log in to unmask]>
Please place this address in your address book
Please purge all others

Web site: Parkinsons Resources on the WWWeb
http://www.geocities.com/murraycharters

----------------------------------------------------------------------
To sign-off Parkinsn send a message to: mailto:[log in to unmask]
In the body of the message put: signoff parkinsn